INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_35.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.35.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

éĹ´

-0.26

å®ģéĿĻ

-0.25

ure

-0.25

çļĦæĺ¯

-0.25

 Programme

-0.25

 Disorder

-0.24

DIG

-0.24

dig

-0.23

ç¡Ģ

-0.23

Ø§Ø³

-0.23

POSITIVE LOGITS

inf

0.27

uple

0.25

AEA

0.25

 metall

0.25

 instruments

0.25

 instrument

0.25

 vinc

0.24

 resumes

0.23

 comet

0.23

 backs

0.23

Activations Density 0.130%

No Known Activations

This feature has no known activations.